Fundam. appl. NemalOl., 1994, 17 (4),357-367

Aspects of resistance in deepwater to the stem nematode Ditylenchus angustus

Richard A. PLOWRlGHT andJoanna R. GILL JnlemalionalJnstit'llteofParasitology, 395a,HatfieldRoad,St. Albans, Herts AJA OXU, Great Britain. Accepted for publication 13 October 1993.

Summary -A seedling based technique for screening for resistance to D. anguslUs in deepwater rice was developed and used to examine components of variabiliry in host suitability. The techniques mimicked natural infection from water and achieved 100 % infection of susceptible check cv. NC492. Approximately 10% of inoculum invaded seedlings, the infective stages being predomi­ nantly J3,J4 and adult. Reproduction of D. angus/Us was rapid on susceptible plants (period oflife cycle was 10-20 days at 30 OC). Symprom expression and nemarode multiplication were studied on a range of resistant and susceptible deepwater rice cvs and lines. Resistance is in part conferred by a rapid necrotic response to feeding in the host. The response was exhibiœd by sorne plants of lines, namely, CNL 319, Bazail65, Rayada l6-02, Rayada l6-03 and Rayada 16-06 ro Rayada 16-09, which, in the field, have been consisrently resistant across locations and years. The response is qualitatively different from the weil known susceptible response and provides a basis for genorypic selection. The relationship between water depth and seedling stature appears to be important in determining infection and early symptom development. D. anguslus invades primarily at the water surface so submergence of the leaf sheath delays infection. Symproms develop more quickly if the warer level during infection is adjacent ro, or just below, the collar at the rop of the leaf sheath. The expression of symproms and hence early damage is delayed in shallower watet.

Résumé -Aspects de la résistance du rizflottant au né'matode des tiges, Ditylenchus angustus - Une technique utilisant les plants de riz a été mise au point pour cribler la résistance du riz flonant à Dllylenchus angus/Us. Cene technique copie l'infestation natureUe à partir de l'eau et permet une infestation de 100 % sur le cv. témoin sensible NC492. Environ 10 % de l'inoculum pénétre dans les racines, les stades infestants étant surtout représentés par lesJ3, J4 et adultes. La reproduction de D. angus/Us est rapide sur les cvs sensibles (la durée du cycle est de 10 à 20 jours, à 30 OC). L'expression des symptômes et la multiplication du nématode ont été étudiées sur une série de cultivars et lignées de riz flonant, tant sensibles que résistants. La résistance est en partie conférée par une réaction nécrotique rapide de l'hôte lors de la prise de nourriture du némarode. Cette réaction est présente chez quelques lignées (CNL 319, Bazail 65, Rayada 16-02, Rayada, 16-03, Rayada 16-06 à Rayada 16-09) qui se SOnt montrées en champ résistantes en divers endroits et au cours du temps. Une teUe réaction est qualitativement différente de ceUe, bien connue, montrée par les plantes sensibles, et peut ainsi fournir Wle base en vue de la sélection génorypique. La relation entre profondeur de l'eau et taille des plants de riz est d'une importance cruciale pour le déclenchement de l'infestation et le développement des symptômes. D. angus/Us pénètre dans les plants d'abord au niveau de la surface de l'eau, aussi une submersion des gaînes des feuilles retardera-t-elle l'infestation. Les symptômes se développent plus rapidement si, durant l'infestation, le niveau de l'eau est voisin de ou trés peu inférieur à celui du collet situé au sommet de la gaîne foliaire. L'expression des symptômes, et par conséquent les dégâts précoces, sont ralentis en abaissant le niveau de l'eau. Key-words : Host-parasite relationships, resistance, screening, Dùylenchus angusLUs, deepwater rice, population dynamics, Iife cycle, Ufra.

The stem nematode of rice, Ditylenchus angusLUs, lam, 1993), D. anguslUs now also infects the lowland which causes the disease known as Ufra, occurs in sever­ transplanted Aman and Bora crops (Miah & MondaI, al Asian countries where it has been predominantly as­ 1988). Although the distribution of the nemarade in sociated with deepwater rice (Bridge el a!., 1990). hs these different crops requires assessment, it is c1ear that pest status arises from its potential, in cornmon with the characteristic pattern of sporadic complete crop fail­ other stem nemarades, ra overrun its host which has ure in endemic areas is oceuring in lowland riee in Ban­ sporadically resulted in complete crop failure, some­ gladesh (M. L. Rahman pers. comm.). times over large areas. As rice cropping practices shift Rice cultivars with resistance to D. anguSlus could away from deepwater rice rawards more productive beeome a central component of Ufra management. Be­ lowland systems, the importance of the deepwater crop ing preventative rather than curative they would be ef­ has been diminished. In Bangladesh, however, where fective against outbreaks of the disease where perhaps, the same shift in cropping practices is taking place (Is- for socioeconomic reasons, other control meas ures are

ISSN 1164-5571/94/03 S 4.00/ © GaUlhier- Viltars - ORSTOM 357 R. A. Plowright & J. R. Gill

not implemented. Measures such as stubble burning, Two of the factors which are likely ta affect initial destroy valuable fodder and chemical protection re­ invasion by D. anguslus, i.e. inoculum infectivity and quires treatment before symptoms are apparent to the water depth, were the subject of this study. farmer. Sources of resistance to D. angusLUs have been identified e.g; Bazai! 65, CNL 319 and the Rayada MATERIALS AND METHODS group of lines (Rahman, 1987) which, despite sorne D. angusLUs was maintained in monoxenic culture on variation across years and locations, have been effective seedlings of lowland rice cv. IR 36 (Plowright & Ake­ against populations of the nematode from , hurst, 1992). AIl the work reported here used a popu­ Bangladesh, and Burma (Anon., 1986, 1987, lation of D. anguslus from Hau Giang Province, Viet­ 1988). These selections, made in pot, deepwater tank nam. Unless stated otherwise' experiments were done in (Rahman & McGeachie, 1982) and field experiments, a heated glasshouse (25-35 oC) with a 10 h minimum were based on assessments of the proportion of stems or photoperiod. When preparing suspensions of inoculum, tillers infected by the nematode, supported in sorne J4 and adult stages of D. anguslus tended to aggregate, cases by counts of nematode populations. The exploita­ particularly if population densities exceeded 500 nema­ tion of resistance requires additional screening tech­ todes/ml. Aggregates were dispersed by prolonged agi­ niques to enable the testing of large numbers of plants. tation. Counting of extremely active nematodes was Techniques are required to characterise genotypes and achieved by cooling, anaesthetising with COz or by trap­ facilitate selection in the early generation screening of ping nematodes in a shallow layer of water in a Bridge breeding programmes. counting dish (Hooper, 1990). For the latter treatment a The broad objective of this work was to provide in­ trace of detergent was used to lower surface tension. formation on the host-parasite interaction to support the Rice seed was provided by the International Rice Re­ development of a higher volume screen for resistance to search Institute. D. angusLUs. The work is presented under two headings. lnfeclivity of slages ofD. angustus The flISt deals with practical aspects of inoculating D. anguslus with associated biological information and the Developmental stage distribution analysis of inocu­ second deals with population dyn:::nlÎcs and variability lum from several experiments has indicated that popu­ in the host-parasite interaction. The discussion brings lations harvested from cultures 40-60 days after inocu­ together infonnation from both areas to enable an as­ lation have a stable demography and comprise eggs sessment of the technique and the components of resist­ (21 %), J2 (7 %), J3 (14 %) J4 (33 %) and adtùts (25 %). ance to D. angusLUs. In order to determine whether equal weight should be given to ail stages in terms of infection potential, the infectivity of each was compared. Examination of the precision Deve10pmental stages of D. anguslus were isolated ofinoculation techniques manually using a low power binocular microscope. Several methods have been used to infect rice with D. Stages were defined from measurements of body length angusLUs, (Hashioka, 1963; Sein, 1977; Rahman & of 246 nematodes (Fig. 1). AdLÙt male and female body McGeachie, 1982; Ou, 1985; Rahman & Evans, 1987). lengths overlapped. Techniques have usualJy mimicked the field condition described by Cox el al. (1980) where secondary in­ 25 J2 J3 J4 Adult oculum is water borne spreading from primary infection loci within the field. This emphasis has been based on 20 the assumption that barriers to, or escape from, infec­ '-- Q) tion may be important components of resistance. In­ .0 E jecting young seedling or leaf sheaths of older tillers :::J c 15 would circumvent these mechanisms. Q) The success of inoculation methods has been assessed "0 9 10 in terms of the proportion of susceptible tiIIers which ~ E become infested. However, to discriminate single plant Q) genotypes within segregating generations or heterogene­ Z 5 ous rice lines it is necessary to have inoculation tech­ niques which enable homogeneous infection and reduce environmental and error components of variation within o-l---r----r-----'-,--r-'----r---r'-r----r-....,-----1 and between lines. Precise control of the initial popu­ 260 340 420 500 580 660 740 820 900 980 1060 lation density/plant (P,) of D. anguslus is vital if rates of Nematode length (~m) reproduction are to be used to separate levels of resist­ ance. Variability in Pi may also influence the assessment Fig. 1. The frequency distribution of 40 /Lm size classes of 246 of variation in host responses of a qualitative nature. venn~r01"m stages ofDirylenchus angustus.

358 Fundam. appl. NemaLOI. Resistance ta Ditylenchus angusrus

Seedlings or rice cv. NC492 were inoculated using a with water of depth 10 cm; 8 : 500 /-LI of inoculum con­ modification ofthe coleoptile method (Hashioka, 1963). fined using a 2 cm diameter tube in water ofdepth 5 cm. Seedlings were inoculated in pairs, when they measured One hundred nematodes and eggs/plant were inoculated 3-5 cm, by introducing 75 nematodes or eggs into 2 ml 12 days after sowing and periodic checks of inoculum of water in 7 ml sterile bijou bottles (Sterilin). Repli­ homogeneity were made in duplicate. Water 1evels were cation was four fold. adjusted and left for 24 h before inoculation to enable Seven days after inoculation (DAI) the bottles were water temperature to equilibrate at 30 oc. filled with boiling acid fuchsin (0.1 % w/v) in glycerol, Seven days after inoculation ail above ground tissue lactic acid and distilled water (1: 1: 1) (Bridge et al.) was harvested and stored in plastic bags at - 20 oc. De­ 1982). The plants were teased in distilled water and frosted plants were stained with hot acid fuchsin counts ofail nematode stages in the seedlings and water (0.1 % w/vl) in equal parts glycerol, lactic and distilled were made. water (Bridge et al., 1982). Stained tissues were cut into Compan·son ofinoculation methods 3-5 mm lengths and macerated in 50 ml of distiiled wa­ The objective of these experiments was to compare ter for 5-8 s. Debris was washed strongly on a 313 /-Lm direct injection methods with methods varying the prox­ sieve with a further 50 ml of distilled water from a hand imity ofinoculum confinement at two depths ofwater at pumped spray. Counts of nematodes were made follow­ inoculation. Plastic drinking straws provided 6 mm dia­ ing settling and large numbers were estimated from du­ meter tubes for confining inoculum close to plants. Each plicate 5 % (v/v) subsamples. straw was cut 4 cm longer than the water depth to allow Influence of water depth at inoculation room for the inoculum volume. Ten, 100 ml pots (vaca­ Five rice cys were selected to provide susceptible and pots, H. Smith Plastics) ofsteam sterilised clay loam soil resistant plants ofdifferent seedling stature, viz. NC492, were placed in each of eight deep boxes tail susceptible, Rama, short susceptible, Hashiamon, (30 x 24 x 17 cm) and sown with ri ce cv. NC492 at the short susceptible, Bazail 65 tail resistant, Rayada 16-06, rate of 1 seed/pot. Each box was assigned one of a num­ short resistant. Seed were sown in 100 ml vacapots in ber of inoculation treatrnents (Table 1) viz: 1 : 5 /-LI of deep trays assigned a water depth of 0, 1, 5, 7_5 or inoculum injected into stems, severed 5 cm above the 10 cm. A randomised block design with four replicates soil; 2: 5 /-LI of inoculum injected between the leaf was used. Twelve days after sowing, seedling height sheath and the emerging new leaf and maintained there­ measured to the coilar at the top of the leaf sheath was after in a saturated environment which was achieved recorded. An inoculum of 250 nematodes was intro­ using an unvented, clear Perspex propagator; 3 : same duced into the volume of water confined by a 6 mm as treatrnent 2 but with diurnal fluctuations in glass­ diameter tube placed around each seedling. Water levels house relative humidity which was monitored for the were again established 24 h before inoculation. Each day duration of the experiment and varied between 60 % after inoculation observations of symptom occurrence and 80 %; 4 : 500 /-LI of inoculum released into water of and type (resistant or susceptible) were made on the depth 5 cm, adjacent to each seedling but not confined, most recently emerged or emerging leaf. The severity of 5 : same as treatrnent 4 but with water depth 10 cm; 6 : susceptible symptoms were also assessed by scoring 500 /-LI of inoculum confined within a 6 mm diameter symptoms according to a scale shown in Fig. 2 (Plow­ tube in water of depth 5 cm; 7 : same as treatrnent 6 but right et al., 1992)_ The number of nematodes/plant was Table 1. The precision ofdijferent techniques for ùwcuiating Di­ determined, as described above, 7 days after inocula­ tylenchus angusrus tion.

Inoculation technique Water Infetted Pi depth plants (cm) (%) mean CV

1 Injection severed shoots 5 100 32 22 2 Leaf sbeath 100 %r.h. 5 100 22 24 3 Leaf sheath 60-80 %Lh. 5 80 2 39 4 Free in water 5 40 1 41 5 Free in water la 80 4 39 6 Confined in 6mm diam_ tube 5 100 8 16 7 Confmed in 6mm diam. tube la \00 6 46 8 Confined in 2cm diarn. tube 5 70 3 42 o 1 2 4 8 16 \-3 : inoculum volume = 5 ftl 4-8 : inoculum volume =500 ftl Fig. 2. A scheme for scon-ng the seven-ty ofsusceptible symplOms of CV = Coefficient of variation infection by Ditylenchus angusrus on nce.

Vol. 17, n° 4 - 1994 359 R. A. Plowright & J. R Gill

RESULTS bur indicared that confinemenr of inoculum in a warer Infectivity depth of S cm was the leasr variable (C of V 16 % Ta­ ble 1). Ali vermiform srages invaded the planr (Fig. 3). The Eifect of zoateT depth on infection proportion of infective J2's thar were either harched from eggs or inoculared was very low and these did nor Ar inoculation, rice cv. NC492 was raller than the moulr. GnlyJ3, J4 and adulr inoculum esrablished repro­ other cultivars (P =O.OS) and the culrivar differences in ducing popularions composed of eggs and ail vermiform seedling srature were consisrenr with the basis of selec­ srages. An estimare of the infectiviry of each srage (the tion although the differences were small (Table 2). As in sum of the inoculared and older srages [up ro adulr] in the previous expetimenr the population in planrs 7 DAI planrs, expressed as a proportion of the number in­ was equivalenr ro ~ 10% of inoculared nurnber. Com­ oculared) indicared thar infectivity \Vas ranked bining warer depth rrearrnenrs indicared thar more ne­ J4 > adulr > J3 > J2 > egg and thar approximarely marodes invaded tice cvs Rama and NC492 than Raya­ 23 % of an inoculared population of vermiform stages in da 16-06 or Hashiamon; Bazail 6S \Vas inrermediare equilibrium is infective. The J4 also had the highesr be[\veen both pairs (P> 0.01). This ranking of culti­ reproductive porential (Fig. 3). Low numbers of J3's vars, however, was nor the same ar each warer depth were found in reproducing popularions suggesring a (culrivar x water depth inreracrion P > O.OS). In general rapid moulr ro J4 and adulr which is further supporred infecrion of cvs NC492, Bazail 6S, Rayada 16-06 and by the predominance of these srages in the demographic Hashiamon was independenr ofwarer depth whilsr more equilibrium of cultures (see above). The presence ofJ4 nemarodes invaded cv. Rama in warer depths of0, 1 and in planrs only 7 days after inoculation with adulrs in­ S cm than ar 7.S and la cm (Table 2). dicares a very short life cycle of approxirnately la days at 30 oc. D Infective proportion Inoculation methods ~ 30 ~ pf/plant A 100 % infecrion of inoculared planrs was achieved E ::::J 160 by direct injection of severed ShONS, leaf sheath inocu­ Oi 25 0 lation and by confining inoculum within a 3 mm radius 0 s of planrs (Table 1). Successful infection by leaf sheath 20 120 inoculation required a high, srable relative humidity. 0 ë c Ci'" Fluctuating humidity, wider or no confinemenr of in­ .'2 15 ...... "- 80 oculum, produced a reducrion in P, (number of nema­ 0 0- n. rodes/planr after 7 days) and the proportion of infecred 2 10 planrs. Direcr injection and leaf sheath inocularion in a n. Q) 40 saturared environrnenr esrablished a higher P, than ~ 5 other rrearrnenrs which mimicked natural invasion from Ü ~ n ~ warer. Leaf sheath inoculation required small (S-10 f..LI) .s 0 0 inoculurn volumes which creared the additional pro­ Eggs J2 J3 J4 Adult blems of inoculum hererogeneity through nemarode ag­ Fig. 3. The relative injectivity ojstages ojDitylenchus angusrus gregation. Coefficienrs of variation were generally high ln nce.

Table 2. Influence oj water depth on the initial injection oj rice by Ditylenchus angusrus

D. anguslus/plant 7 DAI at water depth (cm)

Rice cv/line Host Leaf sheath a 5 7.5 la mean suitability height (cm)

Rayada 16-06 R 7.6 bc 13 ab 7 b 6b 4a 3a 6b Bazai! 65 R 8.0 b la b 5 b 21 a 6a 7a la ab Hashiamon S 6.8 c 7b 15 ab 6b 2a 2a 6b NC492 S 11.0 a 17 ab la b 18 a 4a 12 a 12 a Ratna S 7.0 bc 22 a 24 a 15 ab Sa 1 a 13 a

Dara mean of four replicares. ln a colurnn trearment means having a common letter are nor significantly different, P = 0.05. Hosr suirabûiry R = resisrant, S =susceptible.

360 Fundam. appl. NemalOl. Resistance LO Ditylenchus angusrus

The appearance of symptoms in the young leaf varied maining boxes were harvested at 7 day intervals after with rice cultivar and water depth. With sorne excep­ day 20, giving a total of 18 harvests. At each harvest, tions, more plants had symptoms within 7 days of in­ counts of eggs and ail vermiform stages were made as octÙation, when the water level at inoculation was ad­ above. One plant of each cultivar was eut into 2 cm jacent to, or just below the coUar at the top of the leaf sections, above the peduncle, to examine the distribu­ sheath (Table 3). Combining cultivars, more plants tion of nematodes within the leaf sheath and basal leaf showed symtams in water of depth 7.5 and 10 cm, than regions. at 0 and 1 cm P = 0.05 (Table 3). Although aU the culti­ Variations in host response vars were infected at water depth 0 cm, only Bazail 65 exhibited symptoms within 7 days of inoculation. In Forty predominantly deepwater rice cultivars and Rama and NC492 and ta sorne extent Rayada 16-06 !ines (Table 4) were selected for inclusion in this experi­ symptom expression appeared to be delayed in shallow­ ment. Selection was based on citations of their suscepti­ er water, for example, NC492 showed symptoms 3 bility or resistance to D. anguslUs in the field, although D.A.I. only in 10 cm of water and cv. Rama showed ten untested lines from the 1989 International Ufra symptoms within 24 h in 7.5 cm water but after 5 days screening set were also included. The entries were sown in 5 cm. We have now found that at optimum water in four completely randomised blocks in 100 ml vaca­ depth for symptom expression both resistant and sus­ pots in deep boxes (20 planting stationslbox). Twenty ceptible responses can be expressed within 24 h of in­ plants of a susceptible check entry NC492 were sown at fection. random throughout each block. SeedJings were inocu­ lated with 300 D. angustus, as described above, Il days after sowing and tubes were removed 7 days after in­ Nematode population dynamics, oculation. The plants were examined daily and symp­ syntptomatology and variations in host responses tOms of D. anguslUs infection were scored 7,14,21 and MATERJALS AND 1VŒTHODS 28 days after inoculation using the scale shown in Fig. 2 (Plowright et al., 1992). After the final assessment, ail Population dynamics plants were eut at soillevel and stored in plastic bags at Seeds of resistant cvs Rayada 16-06 and Bazail 65 and - 20 oc. Counts of eggs and nematodes were made in susceptible cv. Habiganj Aman l were sown in vacapots the usual manner. in deep boxes. Fifty seeds were sown in boxes, each In a second experiment examining intravarietal var­ containing a single cultivar, and five seeds of each culti­ iation in host response, 100 seeds of deepwater rice var were sown in a further eight boxes. Ten days after Rayada 16-06 were screened following precisely the sowing each plant was inoculated with 100 D. angustus same procedure. using 6 mm diameter tubes; water depth being adjusted to 7-8 cm. This depth approached the collar at the top of RESUl..TS the leaf sheath in the majority of plants. The tubes were Population dynamics removed 7 DAI. Five randornly identified plants from single cultivar boxes were harvested at 2 day intervals Approximately 10% of the inoculated nematodes in­ foUowing inoculation. Ali the plants in one of the re- vaded seedlings of both resistant and susceptible varie-

Table 3. The influence of waLeT ckpth on aCCU/Tence of syrnptoms in me plants wùhin 7 days of infeaion by Ditylenchus angusrus

Rice cv/line Host Leaf Proportion of plants with symptoms (%) suitability sheath water depth (cm) height (cm) 0 5 7.5 10

Rayada 16-06 R 7.6 bc Da Da Da 25 a 25 a Bazail 65 R 8.0 b 50 ab 25 b 75 ab 100 a 100 a Hashiamon S 6.8 C Da 50 a Da 50 a 25 a NC492 S 11.0 a Ob 25 ab Ob Ob 75 a Ratna S 7.0 hc Ob Ob 75 a 75 a 50 a

Mean 10 20 b 30 ab 50 a 55 a

Host suitabiliry :R =resistant; S =susceptible. Leafsheath heights with a common letter are not signifïcanùy different CP =0.05). In a row, means of symptom occurrence with a common letter are not signifïcanùy different P =0.05.

Vol. 17, n° 4 - 1994 361 R. A. P/owrighl & J. R. Gill

Table 4. Reproduction ofDiryienchus angustus on 40 rice cultivars and lines, 28 days afler ùweulation.

Rice cv/line IRTP ST D. angusLUsl Rice cv/line IRTP ST D. anguslusl No. plant No. plant

Rayada 16-09 13876 R 13 BKNFR 76046-10/2/511/6/0/4 13082 S 588 Rayada 16-11 13867 R 24 BR 425-189-1-6-2-1-2 14902 S 851 CNL 319 06200 R 75 CR 156-5021-207 03199 S 911 Rayada 16-07 13874 R 76 CN 506-147-14-2 11472 S 950 Rayada 16-08 13875 R 80 BKNFR 76046-10-2-5-1-6-0-2 13081 S 998 Bazail65 13861 R 129 Ratna 01007 S 1066 Hashiamon 13864 S 158 IR 36 00266 S 1074 Karkati 161 13866 R 191 NC 492 12789 S 1118 Gowai 50-9 13865 S 217 BKN 6986-66-2 04528 S 1164 Rayada 16-02 13870 R 222 BR 539-83-4-2-2 14918 S 1164 Rayada 16-06 13873 R 236 BKNFR 76046-10-2-5-1-1-0-1 13079 S 1190 BR 425-189-1-6-1-2 14578 S 284 BR 716-7-2-1-1 14574 S 1195 Habiganj Aman 1 06059 S 353 BR 1185-2R-6 14580 S 1196 IR 2307-247-2-2-3 04477 S 374 IR 40905-11-3-3-4-2-21 16122 S 1292 CNL 231BIB 06201 S 408 Rangabao 15057 S 1357 IR 50 07847 S 435 CNM 539 11470 S 1472 Padmapani 15025 S 458 BR 425-189-1-6-2-1-2 14902 S 1578 Rayada 16-03 13871 R 542 IR 40905-1/-3-1-6-3-21 16119 S 1625 PJNB 96-10-1 15053 S 565 BKl\.TFR 76045-85-1-1-1-0-1 13078 S 1665 Cilla 05689 S 585 BR 539-83-4-2-2 14918 S 2002

Data are means of four repetitions. Standard error of difference between IWO means =441 CP =0.01). ST =Symptom type. ties. J3, J4 and adult stages had invaded by day 2, J2's Variaàons in host response were not present until day 8 and were associated with a Symptoms were apparent on the most recent leaf of peak in egg numbers. The dynamics of the population sorne cultivars withill 2-3 days of infection. Symptoms were characterised by a series of rising peaks and on sorne plants of resistant entries, namely, Bazail 65, troughs (Fig. 4) and, within a variety, the timing offluc­ CNL 319, Karkati 161, and the Rayada group of !ines tuations in number of eggs and ail vermiform stages differed from normal Ufra syrnptoms. In these plants a coincided. The period between peaks ofegg production, browrùng response rapidly followed chlorosis ofaffected 10-20 days, was considered to approximate 10 the ne­ tissue and was visible 1-2 days after irùtial symptoms. matode generation time and indicated little difference The response was varied but lOok one or more of three between resistant and susceptible varieties (Fig. 5). Si­ milar proportions of the harvested plants of resistant and 10~ susceptible rice were infected by 8 DAI, but observa­ .eggs tions of ail plants in the experirnent revealed symptoms ... J2 1000 'f' J3 in a greater proportion ofresistant plants than in suscep­ +J4 tible. By 80 DAI more plants of Habiganj Aman 1 re­ .... * Adult c 100 • Total mained infected with D. angusLUs than either of the re­ '" sistant cultivars (Table 5). Nematode numbers ...... 0.. "­ 10 increased more rapidly and to higher levels in the sus­ QI ..c ceptible cultivar CP =0.05) (Fig. 6). The population of E :::J D. angusLUs declined 10 zero in Rayada 16-06 after flo­ Z wering; a small population ofeggs remained in one plant of BazaiJ 65 whilst in Habiganj Aman 1 a population of 0.1 J4's and adults remailled. Nematodes were distributed throughout the leaf sheath above the peduncle, numbers ,I declirùng toward the base and top of the culm and dis­ o 16 32 48 64 80 tributed about a peak density. There is sorne evidence that whilst the distribution of nematodes remained the Days after inoculation same, the position of peak population density migrated Fig. 4. The population dynamies of Dirylenchus angustus on upwards as the plant matured (Table 6). deepwaœr riee cv. Habiganj Aman 1.

362 Fundam. appl. NemalOl. Resistance ta Ditylenchus angusrus

1000 ~ Habiganj Aman ~ Habiganj Aman .., Rayada 16-06 .., Rayada 16-06 _ Sazail 65 ". c: 1000 -Sazail65 Â~'" -cv 100 a...... Cil 100 Cl ~ A\7'~ c: ­cv 10 Cl 1 Q) ~ f\ a. I!---l- '1 \...... ~ l Cil , 10 rLJ~ T 1-~ ... , Cl , Cil .:i Cl , al VjV~.!. UJ ,, "0 ",..,../} \... / T , o ~ '- / \ .. 1 , CV ~ ,)II';, , - : 'i' \ 1 0.1 , E 0.1 ,, \ ,1 fi , al ~ j ,' , : : \ 1 , \ _L." _\ __'- Z _u_ 1. _,...1. .l.ll ,', T , T -n' 1" -'T TTT \1 ~ .u t-,- ''''''I- o +---;.-~ --r---,---__~__._.,._ 0 ... , ..., .. -', • ~ 1 • .-, o 16 32 48 64• 80 o 18 36 54 72 90 Days after inoculation Days after inoculation Fig. 5. The dynamics ofegg populalions of Ditylenchus angus­ Fig. 6. The population dynamics of Ditylenchus angusrus on rus on susceplible deepwaler rice cv. Habiganj Aman 1 and resisl­ susceptible deepwaler rice cv. Habiganj Aman 1 and resistanl cvs anl cvs Rayada 16-06 and BazaiL 65. (Error bars indicate stan­ Rayada 16-06 and BazaiL 65. (Error bars indicate standard dard error of means). error of mean).

Table 5. Differences in infection by DytiJenchus angusrus of Table 6. The dislribution of Ditylenchus angusrus within lhe resistanl and susceptible deepwaler rice. Leaf shealh ofdeep waler rice priOl" LO lhe onset offlowering.

Days after Population range and peak Rice cv/line Host Infected plants Plants with infection suitability (%) symptoms density (%) height above penducie (cm) 8 DAI* 80 DAI** 8 DAI*'" Range Peak Bazail65 R 56 35 91 8 2-6 4 Rayada 1&-06 R 47 35 53 14 0-12 2 Habiganj Aman 1 S 65 80 28 20 8-12 8 27 0-12 6 Host suitability :R: resistanr, S: susceptible, from field assessments 34 0-18 8 '·n =20 observations 41 0-16 6 '''i'n 90 observations = 48 0-20 10 55 0-22 12 forms: i) Browning of the penultimate leaf base; il) Browning of the midrib, sometimes along its entire length; iii) Browning within a more or less discrete, yel­ were symptomless and 44 % were scored 4, although at lowish halo on the leaf lamina (Fig. 7). Subsequent leav­ 30 DAI 97 % of plants were scored 8 or 16 and none es were often symptomless. Occasionally the response escaped infection. In sorne cultivars, for example Hash­ was noted in later !eaves, both in plants which had previ­ iamon and Gowai 50-9, which exhibited a susceptible ously exhibited the symptom or those that had been response, the mean reproduction of D. angustus was symptomless. Sorne plants of resistant entries remained equivalent to that of sorne resistant lines (Table 4). symptomless throughout the 30 day experiment. Sorne plants of ail the resistant entries supported re­ Ail plants of the susceptible entries exhibited symp­ production of D. angustus e.g. one replicate plant of toms, but the speed of development and severity of Bazail 65, CNL 319 and Karkati 161 supported popu­ symptoms varied berween entries. For example 7 DAI, lations of 450, 300 and 735 D. angustus respectively. A symptoms on cv. Rama had reached a severity of 8, similar variation was evident amongst the Rayada group whilst symptoms on cv. Habiganj Aman 1 required and Il of 28 plants supported a multiplication factor of 30 days to reach the same severity and were symptom­ > 2. Within a population of 100 plants of Rayada 16-06 less 7 DAI. There was slight variation berween plants of individuals displayed either resistant (49 %) or suscep­ the susceptible check cv. NC492. Seven DAI, 25 % tible (38 %) responses or were symptomless (13 %) the

Vol. 17, n° 4 - 1994 363 R. A. Plownghl & J R. Gill

Fig. 7. Responses of deepwaler rice la infeClion by Ditylenchus angusrus. A : SuscepLible : B-D: Resislant : noie necrosis of mid-rib and pale green-while patehes on lhe leaf blade somelimes wùh a reclangular appearance. In high magnifualion ofleaf blade (D) noie pale green MW (J) around discrele yelww-white area (2) whuh precedes necrosis (3). response being consistent with the presence and number 2000 • of nematodes (Table 7). Within the susceptible group, r=0.735, P<0.001 symptom severity rating 28 DAI was linearly correlated 1 with number of D. angus/Us (r =0.697, P < 0.001). No ::: 1500 such correlation was found in the resistant group. a... • ...... • Scores of symptom severity for ail entries made ci • 28 DAI, were linearly correlated with number of D. an­ z • • • gustus/entry (r = 0.735, P < 0.001) (Fig. 8). The rela­ '" 1000 • • .:! • tionship did not hold for symptoms assessed at earlier "'" • rimes. Analysis of the mean number of D. angustus/plant .."'" of ail entries indicated significant differences ci 500 -. • (P = 0.001) (Table 4). Variations among resistant en- •• • • ".. • • Table 7. Varialions of hOSI response of deep waler rice, Rayada o 4 8 12 16 J6-06 LO infeClion by Ditylenchus angusrus. Symptom severity score

Host response Proportion of D. anguslus Fig. 8. The relalionship belween symplOm severùy and reproduc­ plants ppplant lion of Ditylenchus angusrus in deepwaler rice 28 days after inoculaLion. Symptorrùess 13 o Resistant 49 2 (0-13) tries were not significant, but among susceptible entries Susceptible 38 46 (6-572) (resistant entries removed from the analysis) there were differences in the level of susceptibiliry (P =0.01).

364 Fundam. appl. NemalOl. Resistance LO Ditylenchus angustus

The coefficient of variation of susceptibility for the ceptible cv. Habiganj Aman l (Fig. 6). To sorne extent susceptible check was high (40-50 %) which enabled nematode reproduction on resistant entries can be ex­ differentiation of levels of susceptibility differing by a plained by genotypic variability within a population of multiplication factor (Pr:actual P) ofapprox. 29-30, i.e. plants, for example, the Rayada lines comprise resistant approx. 900 nematodes. Estimates of error variance and susceptible genotypes (Table 7). The isolation of suggest that 10 fold replication would be required to nematodes in necrotic host lesions is involved in resist­ reduce the Least Significant Difference to 300 nema­ ance to Dùylenchus dipsaci. But, it is also possible to todes. envisage how such a necrotic reaction might fail to com­ pletely restrict an active ectoparasite such as D. angus/us Discussion and thus allow sorne reproduction. The principal concern for resistance screening is the The screening techniques used in this work revealed extent to which genotypic, environmental and error significant differences in resistance to D. angus/us (de­ sources contribute to the observed variability. Answers fined solely in terms of the influence of the host on to these questions indicate important components of nematode reproduction, see Cook and Evans, 1987) resistance. Escapes due to inoculum failure are a pos­ between cultivars and lines of deepwater rice. In com­ sible source of error, particularly when mimicking natu­ mon with other stem nematode host-parasite relation­ rai infection of rice by D. angus/us. The proportion of ships (Bingefors, 1970; Williams, 1972; Cook & Evans, inoculum invading the plant was consistently low, usu­ 1988; Stanton el al., 1984; Whitehead el al., 1987), ally around 10 % of :% 300 inoculated individuals and, there was a good correlation between symptom type and although increasing inoculum sizes would reduce the severity and the number of nematodes infecting the risk of errors from escapes, it would be wasteful of ne­ plant (Fig. 8). Resistant plants were symptomless or ex­ matodes. Achieving P,'s higher than 30/plant is probably hibited a response involving rapid necrosis (Fig. 7). The unnecessary in view of the very rapid life cycle of D. prevalence and severity of symptoms which precede angus/us. Variations caused by differences in the demog­ browning is initially greater in resistant plants (Table 5) raphy of populations of D. angus/us between experi­ which suggest that the response is of a hypersensitive ments are likely to be slight, but populations composed nature. The current whole plant studies precluded ob­ of unusually high proportions of the less infective stages, servations of new leave until they emerged from the leaf viz. eggs,}2 and}3 (Fig. 3), should not be used. sheath so the time scale of events is imprecise. Sub­ The relationsh.ip benveen plant stature and water sequent studies (Plowright & Gill, unpubl.) have in­ depth appears to be important in detennining infection dicated that the response has a growth cost which is and early symptom development. Table 2 suggests that particularly obvious in very young seedlings infection of rice by D. angus/us is reduced if the leaf « 10 Days). The consistently resistant lines from field sheath is completely submerged. It follows that the ne­ trials e.g. CNL 319 and the Rayada lines exhibit this matodes invade primarily at the water surface. Symp­ necrotic response to feeding by D. anguslUs in glass­ tom development is more rapid if the water level coinci­ house tests and it is reasonable to assume that resistance, des with the top of the leaf sheath and is delayed in at least in part, is derived from this host response. Symp­ shallower water. The àppearance of symptoms at the tomless plants are escapes, which have avoided infection base of the leaf blade in rice at aU growth stages points to either through a failure of technique or as a consequence a preference for tissue with meristematic activity. In­ of a disease avoidance strategy. The occurrence of tercalary meristematic activity occurs in this region symptomless plants harbouring D. angus/us has been (Esau, 1965), and probably explains the rapidity of reported (Rahman & McGeachie, 1982; Rahman & symptom development. Subsequent studies (Plowright Evans, 1987), but we have never found D. angus/us in & Gill, unpubl.) have confirmed these observations and plants which remain symptomless throughout the demonstrated that if infection takes place after tillering 28 day period of assessment. has commenced then symptom development is more In general, low populations of D. anguslUs developed rapid in the younger, shorter, tiller providing it is not on entries exhibiting the resistant response. However submerged. Severe damage can lead to death of these benveen the extremes of resistance and susceptibility younger tillers. Disease escape then, can play a role in sorne entries exhibiting resistanr and susceptible re­ resistance to D. angus/us. In the field, however, assum­ sponses supported similar \evels of reproduction (Ta­ ing that nematodes become active for widespread in­ ble 4) and showed no differences in the time required fection at flooding, escape mechanisms would be sensi­ for the nematode to complete a life cycle (Fig. 5). Com­ tive to the interrelationships of sowing date, plant parable levels of susceptibility disguise differences in the growth rate and the onset and rate of flooding. At a given rate of symptom development and, for example, al­ stage of maturity, warer depth determines which tillers though mean numbers of D. angus/us were the same on become infected and the severity of the initial damage to Rayada 16-06 and Habiganj Aman l at 58 DAI, the rate those tillers. Premature flooding could circumvent es­ of population increase was initially greater on the sus- cape mechanisms and result in early damage of crucial

Vol. 17,no4-1994 365 R. A. Plowright & J. R. Glll

importance in establishing infection and determining No. PHF 169/13 and funded by the UK Overseas Devel­ subsequent crop Joss. opment Administration. Barriers to infection may play a role in resistance. References ülder plants are known ta be less easily infected (Rah­ man & Evans, 1987). Among the few rice cultivars and ANON. (1986). Final report of the first international rice ufra lines examined in detail in this work (Table 2), there screening sel 1985. International Rice Research Institure, Phi­ were no consistent differences in the initial infection of lippines, 5 p. resistant and susceptible seedlings. Likewise, similar ANON. (1987). Final report of the second international rice ufra proportions of resistant and susceptible plants initiaUy screening set. International Rice Research Insti tute, Philip­ become infected, the reduction in the proportion of in­ pines, 4 p. fected resistant plants (Table 5) being expressed post­ ANON. (1988). Final report of the third international rice ufra infectionaUy. screening set 1987. International Rice Research Institure, Phi­ From the point of view of screening, seedling and lippines, 5 p. young plant responses agree closely with field assess­ BINGEFORS] S. (1970). Resistance against stem nematodes mems made of the host suitability of mature plants and Dùylenchus dipsaci (Kühn) FiJipjev. EPPO Publ. Ser A.] 54 : provide a strong basis for genotypic selection. Poor cor­ 63-75. relations between assessments of levels of susceptibility BRlDGE,}., Luc, M. & PLOWRlGHT] R. A. (1990). Nematode made in successive screening trials and the sensitiviy of parasites of rice. In: Luc, M.] Sikora, R. A. & Bridge, ]. these assessments to inoculation technique, appear ta be (Eds). Plant Parasùic nemalOdes in subtropical and tropIcal a feature of screens for resistance to stem nematodes e.g. agriculture. WaUingford] UK, CAB InternationaJ : 69-108. (Williams, 1972). Such variability has been a feature of BRlDGE,}., PAGE] S. L.]. &]ORDAN, S. (1982). An improved our studies and is also evident from field screening re­ method for staining nematodes in roots. Rep. Rothamsted sults. Rice cv. NC492, which is a susceptible cultivar, exp. Sin 1981] Part 1: 171. was among the best resistant entries in field trials in 1987 (Anon., 1988). This variability is also expressed in the COOK, R. & EVfu'lS, D. R. (1988). Observations on resistance in white clover (Tnfolium repens L.) to the stem nematode sporadic nature of crop losses (Bridge et al.] 1990). Dif­ (Dùylenchus dlpsaei [Kühn] FiJipjev). Agric. Sei.] 110 : 145­ ferences in seedling stature and rate of growth make it 154. difficu)t ta optimise water depth for nematode infection and symptom development for mixed entries in a COOK] R. & EVANS, K. (1987). Resistance and tolerance. In : screen. It follows that it is equally difficult to conduct Brown] R. H. & Kerry, B. R. (Eds). Pn·neiples and praaise of successive screening trials in precisely the same manner nemalOde control in crops. Academic Press, Australia : 179­ 231. because of temporal fluctuations in plant growth rates. Variability due to differences in seedling stature can be Cox, P. G.] RAHMAN, M. L. & HANNAN] M. A. (1980). minimised by inoculating younger seedlings where the Structura] anaJysis of the nematode population and source height of the leaf sheath and water depth is ;" 5 cm. of ufra. Int. Rlce Res. Newsl., 5 : 8-9. Repeated inoculations by direct inoculation would be ESAU, K. (1965). Plant analOmy. New York] London & Syd­ required to effectively eliminate this source of variabil­ ney, John Wiley & Sons, xx + 767 p. ity. HASHIOKi\] Y. (1963). The rice stem nematode Dùylenchus In conclusion, resistance in deepwater rice is deter­ angustus in . FAO Pl. Protea Bull., 11 : 97-102. mined, in part, by a postinfectional response (probably HOOPER, D.]. (1990). Extraction and processing of plant and of a hypersensitive nature) and seedling growth rate. soil nematodes. In: Luc] M., Sikora] R. A. & Bridge] J. Selection of the former is possible through seedling (Eds). Plant parasùic nemalOdes in subtropical and tropical based techniques which provide a reliable and field rele­ agriculture. Wallingford] UK CAB International: 45-68. vant means of increasing the volume of screening pro­ ISLfu\1] Z. (1993). Seed scarcity and rapid extinction of deep­ grammes. The techniques should be adapted, probably water rice (DWR) cultivars in Bangladesh. In!. Priee Res. by using repeated, direct inoculations] to eliminate es­ Notes] 18: 5. capes. In this work] symtomless plants have been treated MIAH, S. A. & MONDAL, A. H. (1988). Nematodes in deep­ with caution because of their occurrence in known sus­ water rice. Proc. 1987 im. Deepwater Rice Worksh., I.R.R.I. ceptible cultivars. Theil: occurrence in resistant lines is Philippines: 575-582. the subject of further investigation of the mechanisms involved. Ou, S. H. (1985). Rice dlseases. Kew] UK, Commonwealth Mycologicallnstitute, xi + 380 p. Acknowledgements PLOWRIGHT, R. A. & AKEHURST (1992). Monoxenic culture We are grateful to Ms. T. Akehurst for rechnical assistance of the Ufra nematode Dùylenchtls angusws. Fundam. appl. in some aspects of this work and Dr. D.]. Hum for helpfuJ NemalOl., 15 : 327-330. discussions. The work was done as part ofa Narural Resources PLOWRIGHT] R. A. & GILL, J R. & AKEHURST] T. E. (1992). Instirure extramural contract X0094 under Nl.AFF Licence Assessment of resistance and susceptibiJiry ta stem nema-

366 Fundam. appl. NemalOl. Resistance LO Ditylenchus angusrus

tode Dùylenchus anguslus. lnl. Rice Res. News., 17 (3) : 11­ SEIN, T. (1977). Varierai resisrance ra Ufra disease in Burma. 12 & Erratum 17 (4): 23. lnl. Rice Res. Newsl., 2: 3.

RAHMAN, M. L. (1987). Source of Ultra resisranr deepwater STANTüN,]. M., FISHER, ]. M. & BRlTTüN, R. (1984). Re­ rice. lm. Rice Res. Newsl., 12 : 8. sisrance of cultivars of Avena saliva to, and hosr range of, an oar-anacking race of Dilylenchus dipsaci in Soum Australia. RAHNIAN, L. & MCGEACHlE, I. (1982). Screening for resisr­ Ausl. J. exp. Agne. & anim. Husb., 24 : 267-271. ance to Ufra disease (Dilylenchus anguslUs) in deepwater rice. WHlTEHEAD, A. G.) FRASER, ]. E. & NICHüLS, A.]. F. Proc. 1981 im. Deepwaler Rice Workship. !.l.R.I., Philippines: (1987). Variation in me developmenr of stem nematodes, 459-466. Dilylenchus dipsaci, in susceptible and resisranr crop planrs. Ann. appl. Biol., 111 : 373-383. RAHMAN, M. L. & EVANS, A. A. F. (1987). Srudies on me hosr-parasite relationships of rice stem nematode, Dùylen­ WiLLIAMS, W. M. (1972). Laboratory screening of white cla­ chus angusLUs (Nematoda : Tylenchida) on rice, Oryza sali­ ver for resisrance to stem nematode. N. Z. JI agnc. Res., 15 : va, L. NentaLOlogù;a, 33 : 451-459. 363-370.

Vol. 17, n° 4 - 1994 367